The goal of this independent scientist award (K02) application is to provide the Candidate with protected time to conduct research on initiation and progression of alcohol use. A host of powerful analytic methods and readily available specialized software have recently been developed that allow for empirical testing of hypotheses of change, but recent emphasis on statistical advances may divert attention from theory. Making inferences about age-related changes from empirical data requires attention to issues of design, measurement, and analysis. To the degree that the issues affect the correspondence between statistical models and theories of development, our ability to validly test developmental theory is threatened. This proposal seeks to illustrate and address key challenges that arise when translating theory to empirical findings by applying newly emerging methods for modeling development. These include developmental (dis)continuity, heterotypic continuity, approaches to modeling covariates, origin and metric of the time scale, and cohort sequential design. Developmental methods will be applied to four prospective datasets on underage alcohol use: three secondary datasets that span adolescence through young adulthood and the Candidate's NIAAA-funded study, Initiation and Progression through Early Drinking Milestones in Underage Drinkers (R01 AA016838), a study exploring characteristics of the early drinking career in a multi-cohort sample of 1,023 Rhode Island middle schoolers, for which data collection is ongoing and data processing is underway. The secondary datasets, the National Longitudinal Survey of Youth (NLSY97), the Michigan Longitudinal Study (MLS), and the Minnesota Twin Family Study (MTFS), offer advantages with regard to sampling, follow-up interval and duration, operationalization of alcohol involvement, and type of risk factors assessed. Together, these four datasets will address important substantive research questions about the predictive utility of drinking precursors and risk factors at the level of the individual social context, and broader context, as well as outcomes of risky alcohol use. In addition, the extent to which initiation and progression vary by age, across stage of alcohol involvement, and over important life transitions will be examined. Career development activities take the form of research collaborations with nationally recognized consultants in the fields of quantitative psychology and developmental psychopathology that regularly develop, adapt, and employ methods to address questions of development using prospective data. These collaborations will result in joint manuscripts and will provide support for extensive, iterative interactions. Formal training activities include workshop attendance, didactic readings, and dissemination to the field in the form of review papers, symposia, and topical meetings. By the end of the award period, the Candidate will have the skills to bridge state-of-the-art emerging analytic techniques with developmental theory to substantially advance our knowledge of the etiology of alcohol initiation and progression of use and will be a leading expert in the application of models of change.